Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286539743> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4286539743 endingPage "8421" @default.
- W4286539743 startingPage "8409" @default.
- W4286539743 abstract "Partial label learning (PLL), an important branch of weakly supervised learning, addresses the problem that each instance is associated with a set of candidate labels and only one is correct. In this paper, a novel adversarial model PL-MIGAN is proposed to simultaneously mitigate two fundamental issues of generative adversarial networks (GANs) in PLL: label disambiguation performance of discriminator and instance synthesis quality of generator. First of all, multi-class support vector machines (SVMs) applied in discriminator to disambiguate the candidate labels and identify fake instances. This strategy not only improves the disadvantage that traditional supervised loss is unable to perform disambiguation but also reduces the influence of cumulative error caused by noise label propagation. Furthermore, a partial contrastive loss is constructed to extend the self-supervised contrastive approach to PLL, allowing us to effectively leverage ambiguous labels information. Finally, the generator jointly employ mutual information (MI) and partial contrastive loss to estimate the latent distribution of each class label. In addition, in order to reduce the impact of ambiguous information, an iteratively optimization procedure is designed to update the label confidence matrix as conditional information guides the generation of instance classes. As adversarial learning proceeds, both the discriminator and the generator alternately and iteratively boost their performance. Simulation results reveal the overwhelming performance of PL-MIGAN." @default.
- W4286539743 created "2022-07-22" @default.
- W4286539743 creator A5019689779 @default.
- W4286539743 creator A5047282005 @default.
- W4286539743 date "2022-12-01" @default.
- W4286539743 modified "2023-10-05" @default.
- W4286539743 title "Partial Label Learning via GANs With Multiclass SVMs and Information Maximization" @default.
- W4286539743 cites W2024328138 @default.
- W4286539743 cites W2096072446 @default.
- W4286539743 cites W2106008047 @default.
- W4286539743 cites W2112796928 @default.
- W4286539743 cites W2137917285 @default.
- W4286539743 cites W2138621090 @default.
- W4286539743 cites W2142863987 @default.
- W4286539743 cites W2143036568 @default.
- W4286539743 cites W2158140971 @default.
- W4286539743 cites W2338068721 @default.
- W4286539743 cites W2393384312 @default.
- W4286539743 cites W2733555913 @default.
- W4286539743 cites W2808448059 @default.
- W4286539743 cites W2903641274 @default.
- W4286539743 cites W2905443329 @default.
- W4286539743 cites W2951821459 @default.
- W4286539743 cites W2964666459 @default.
- W4286539743 cites W2967363906 @default.
- W4286539743 cites W2987191133 @default.
- W4286539743 cites W2997568094 @default.
- W4286539743 cites W3014734617 @default.
- W4286539743 cites W3036435328 @default.
- W4286539743 cites W3093591919 @default.
- W4286539743 cites W3099156605 @default.
- W4286539743 cites W3160687063 @default.
- W4286539743 cites W3166279462 @default.
- W4286539743 cites W3169510026 @default.
- W4286539743 cites W3172516059 @default.
- W4286539743 cites W3173151551 @default.
- W4286539743 cites W4206196317 @default.
- W4286539743 cites W4220837072 @default.
- W4286539743 cites W4250739957 @default.
- W4286539743 cites W4251481993 @default.
- W4286539743 doi "https://doi.org/10.1109/tcsvt.2022.3192907" @default.
- W4286539743 hasPublicationYear "2022" @default.
- W4286539743 type Work @default.
- W4286539743 citedByCount "1" @default.
- W4286539743 crossrefType "journal-article" @default.
- W4286539743 hasAuthorship W4286539743A5019689779 @default.
- W4286539743 hasAuthorship W4286539743A5047282005 @default.
- W4286539743 hasConcept C119857082 @default.
- W4286539743 hasConcept C121332964 @default.
- W4286539743 hasConcept C153083717 @default.
- W4286539743 hasConcept C153180895 @default.
- W4286539743 hasConcept C154945302 @default.
- W4286539743 hasConcept C163258240 @default.
- W4286539743 hasConcept C2779803651 @default.
- W4286539743 hasConcept C2780992000 @default.
- W4286539743 hasConcept C41008148 @default.
- W4286539743 hasConcept C62520636 @default.
- W4286539743 hasConcept C76155785 @default.
- W4286539743 hasConcept C94915269 @default.
- W4286539743 hasConceptScore W4286539743C119857082 @default.
- W4286539743 hasConceptScore W4286539743C121332964 @default.
- W4286539743 hasConceptScore W4286539743C153083717 @default.
- W4286539743 hasConceptScore W4286539743C153180895 @default.
- W4286539743 hasConceptScore W4286539743C154945302 @default.
- W4286539743 hasConceptScore W4286539743C163258240 @default.
- W4286539743 hasConceptScore W4286539743C2779803651 @default.
- W4286539743 hasConceptScore W4286539743C2780992000 @default.
- W4286539743 hasConceptScore W4286539743C41008148 @default.
- W4286539743 hasConceptScore W4286539743C62520636 @default.
- W4286539743 hasConceptScore W4286539743C76155785 @default.
- W4286539743 hasConceptScore W4286539743C94915269 @default.
- W4286539743 hasFunder F4320322108 @default.
- W4286539743 hasIssue "12" @default.
- W4286539743 hasLocation W42865397431 @default.
- W4286539743 hasOpenAccess W4286539743 @default.
- W4286539743 hasPrimaryLocation W42865397431 @default.
- W4286539743 hasRelatedWork W2789004835 @default.
- W4286539743 hasRelatedWork W2790091315 @default.
- W4286539743 hasRelatedWork W2953246223 @default.
- W4286539743 hasRelatedWork W3035192562 @default.
- W4286539743 hasRelatedWork W3093666952 @default.
- W4286539743 hasRelatedWork W3101578490 @default.
- W4286539743 hasRelatedWork W3139414107 @default.
- W4286539743 hasRelatedWork W3190733223 @default.
- W4286539743 hasRelatedWork W4280544492 @default.
- W4286539743 hasRelatedWork W4293320219 @default.
- W4286539743 hasVolume "32" @default.
- W4286539743 isParatext "false" @default.
- W4286539743 isRetracted "false" @default.
- W4286539743 workType "article" @default.